Separation of potential field data using 3-D principal component analysis and textural analysis

Details Separation of potential field data using 3-D principal component analysis and textural analysis Separation of potential field data using 3-D principal component analysis and textural analysis

Dec 2009

adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2009geoji.179.1397z&link_type=abstract

Geophysical Journal International, Volume 179, Issue 3, pp. 1397-1413.

Astronomy and Astrophysics

Astronomy

1

Image Processing, Spatial Analysis, Gravity Anomalies And Earth Structure, Magnetic Anomalies: Modelling And Interpretation

Scientific paper

Potential field data represent the superposition of effects of all surface and underground sources. A reliable interpretation of different gravity or magnetic anomalies greatly depends on a reasonable separation between regional field and local anomalies. We present here a novel separation method based on a 3-D principal component analysis (PCA) and textural analysis. The PCA, used to decompose the potential field data into a linear superposition of eigenimages, is performed not only on anomaly values but also on textural features, so as to fully use the spatial distribution characteristics of the data and make the separated regional field comprehensively account for the major variations of the data. In order to reduce subjectivity and inaccuracy, we propose a texture-based criterion in separation result selection, which measures the highlighted differences between the two kinds of anomaly by textural statistics and select the first several eigenimages corresponding to the most important variability as the region field when the differences reach maximum. The method is tested with two synthetic models and two real data examples from the Huanghua area, located in Hebei Province, China. Our tests suggest that the method provides a better separation of regional and local anomalies than does the polynomial fitting technique. The separated regional fields and local anomalies of the gravity and magnetic data coincide well with the geological structure of the Huanghua area.

Affiliated with

Also associated with

No associations

Rating

Separation of potential field data using 3-D principal component analysis and textural analysis does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Separation of potential field data using 3-D principal component analysis and textural analysis, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Separation of potential field data using 3-D principal component analysis and textural analysis will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-1887661